Glioblastoma multiforme (GBM) is the most common and aggressive brain cancer with a dismal prognosis and a 3.3% 5-year survival rate. In the past decade, in vitro and in vivo studies have demonstrated that neural stem cells (NSCs), neural progenitor cells (NPCs), and mesenchymal stem cells (MSCs) possess an inherent tropism toward invasive malignancies within the brain, including GBM. These stem cells track down cancer cells in the tumor mass as well as migratory cancer cells in the brain parenchyma. While several of these reports suggest that both stem cell types can be used as a diagnostic tool for gliomas and/or as a delivery vehicle for glioma therapy, recent studies demonstrate several advantages of NSCs over MSCs. The disadvantage of NSCs is the lack of an adequate NSC source and approaches to produce NSCs by induced pluripotent stem cell (iPSC) technologies are fraught with incalculable risks. Thus, identification of easy, fast, and safe methods for production of patient specific NSCs and NPCs and further modification of these cells to increase their tropism to gliomas could lead to the generation of highly specific and effective tools for diagnosis, visualization, and treatment of gliomas. In our previous studies we have demonstrated that MSCs derived from human adipose tissue (hAMSCs) exhibit high migratory ability towards gliomas in vitro, and this tropism can be further enhanced by preconditioning these stem cells with glioma conditioned media (GCM). In addition, the delivery of a therapeutic agent such as bone morphogenetic protein 4 (BMP4) by hAMSCs has been shown to terminally differentiate brain tumor initiating cells (BTICs), the culprits of GBM relapse and metastasis, and significantly prolong survival. These data prompted us to elucidate whether NPCs could acquire even greater and more specific tropism to gliomas when preconditioned with GCM and if they could deliver the extraordinary therapeutic modality that is BMP4. To this end, NPCs generated from human bone marrow MSCs (BM-hMSCs) or hAMSCs via a chemical approach have been used. We found that chemically induced NPCs (CiNPCs) derived from hAMSCs exhibited higher migratory capacity to GCM compared to hAMSCs, and preconditioning with GCM improved their migratory ability. Thus, CiNPCs produced by our recently developed safe, fast, reproducible, and cost-effective cell reprogramming approach could be an ideal BMP4 delivery vehicle for glioma therapies. The goals of this SBIR phase I proposal are to investigate the tropism of GCM preconditioned and lentiviral-BMP4 transduced CiNPCs (LV-BMP4-CiNPCs) to the D54-MG cell line and two BTIC glioma cell lines in vitro (Specific Aim 1), and to test whether they can be utilized as vehicles for delivery of BMP4 to intracranial human glioma xenografts in rodents (Specific Aim 2). Phase II studies will investigate the therapeutic efficacy of LV-BMP4-CiNPCs on different types of gliomas. Commercial and clinically compatible research products emerging from Phase I/II work include technology for large-scale clinical grade production of LV-BMP4-CiNPCs that possess specific tropism to gliomas.